[Talk-us] Peculiar values for natural key in California

[Jerry Clough - OSM]

 I have just been looking through the long tail of natural values and 
natural=K2156 stuck out like a sore thumb. These seem all to be nodes imported 
around 2009 roughly around Salinas : http://overpass-turbo.eu/s/9hM.
I'd appreciate if someone more local could take a look at these and sort them 
out.
Jerry Clough___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] New York, Ellis Island Boundary

[Jim McAndrew]

This is correct, the island is part of NY, but the water is part of NJ,
when they made the island bigger, the new land was in NJ.

Similar to how Delaware has the entire Delaware Bay in its boundaries. They
added some fill to the NJ side, and now there's a piece of Delaware
connected to New Jersey:
http://www.openstreetmap.org/#map=14/39.6142/-75.5635
http://en.wikipedia.org/wiki/Finns_Point


On Sun, May 10, 2015 at 10:35 AM, Luis Villa  wrote:

> Possibly useful context:
> https://en.wikipedia.org/wiki/New_Jersey_v._New_York
>
> On Sun, May 10, 2015 at 9:30 AM Richard Welty 
> wrote:
>
>> On 5/10/15 11:57 AM, Richard Welty wrote:
>> > On 5/10/15 11:31 AM, Frederik Ramm wrote:
>> >> Hi,
>> >>
>> >>puzzled about
>> >>
>> >> http://www.openstreetmap.org/way/37573502#map=18/40.69986/-74.03910
>> >>
>> >> is this really part of today's political boundary, then "(historical)"
>> >> should perhaps be removed from the name. And is there any significance
>> >> to the funny shape (an owl sitting on a branch?) or should it rather be
>> >> aligned with the coastline?
>> >>
>> > this looks like a carve out, a tiny enclave of NYS/NYC surrounded by
>> > New Jersey. i'll look at the newer TIGER data in a little bit and see
>> > what i can see there.
>> now that i have thought about it, Ellis Island is mostly fill. the
>> political
>> boundary likely matches the original waterline of the island and it
>> was simply never changed. as it was operated by the Federal government,
>> there was probably little motivation to do so.
>>
>> richard
>>
>> --
>> rwe...@averillpark.net
>>  Averill Park Networking - GIS & IT Consulting
>>  OpenStreetMap - PostgreSQL - Linux
>>  Java - Web Applications - Search
>>
>>
>> ___
>> Talk-us mailing list
>> Talk-us@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk-us
>>
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us
>
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] New York, Ellis Island Boundary

[Serge Wroclawski]

I agree, the historical boundary should be removed, but we need to be
sure to show what's in what state. It's quite a little mess.

- Serge

On Sun, May 10, 2015 at 11:31 AM, Frederik Ramm  wrote:
> Hi,
>
>puzzled about
>
> http://www.openstreetmap.org/way/37573502#map=18/40.69986/-74.03910
>
> is this really part of today's political boundary, then "(historical)"
> should perhaps be removed from the name. And is there any significance
> to the funny shape (an owl sitting on a branch?) or should it rather be
> aligned with the coastline?
>
> Bye
> Frederik
>
> --
> Frederik Ramm  ##  eMail frede...@remote.org  ##  N49°00'09" E008°23'33"
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us

___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] New York, Ellis Island Boundary

[Richard Welty]

On 5/11/15 11:55 AM, Serge Wroclawski wrote:
> I agree, the historical boundary should be removed, but we need to be
> sure to show what's in what state. It's quite a little mess.
>
>
i think the only problem is that the word historical is there. deleting
the tags on the way would be sufficient. the way itself is shared by a
lot of boundary relations and should not be removed.

i have it loaded up in JOSM right now, but will refrain from uploading
changes for the moment. i'll pull the trigger if there is a consensus.

richard

-- 
rwe...@averillpark.net
 Averill Park Networking - GIS & IT Consulting
 OpenStreetMap - PostgreSQL - Linux
 Java - Web Applications - Search




signature.asc
Description: OpenPGP digital signature
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Peculiar values for natural key in California

[Bryce Nesbitt]

On Mon, May 11, 2015 at 3:17 AM, Jerry Clough - OSM 
wrote:

> I have just been looking through the long tail of natural values and
> natural=K2156 stuck out like a sore thumb. These seem all to be nodes
> imported around 2009 roughly around Salinas :
> http://overpass-turbo.eu/s/9hM.
>

That's a Tiger "feature class" of "Government Center".
http://www2.census.gov/geo/pdfs/maps-data/data/tiger/tgrshp2009/TGRSHP09AF.pdf
http://taginfo.openstreetmap.com/search?q=K2165#values

It's a bad import, but a small one.
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] New York, Ellis Island Boundary

[Frederik Ramm]

Hi,

On 05/10/2015 06:35 PM, Luis Villa wrote:
> Possibly useful context:
> https://en.wikipedia.org/wiki/New_Jersey_v._New_York

That's one of the things that makes me love this hobby/profession so
much. Quirks!

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frede...@remote.org  ##  N49°00'09" E008°23'33"

___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

[Talk-us] Duplicate nodes in Delaware

[Stellan Lagerström]

User toleary made some kind of upload last week of 200k+ nodes in 
Delaware, many of which seem to be duplicates of existing nodes:

http://www.openstreetmap.org/user/toleary/history#map=9/39.0874/-75.5759

Have sent message to user, but without answer.

/Stellan


___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Duplicate nodes in Delaware

[Bryce Nesbitt]

Looks pretty bad.
Bare untagged nodes on top of existing way nodes.
A speedy revert is in order: are you comfortable doing that, or do you want
help?

On Mon, May 11, 2015 at 2:07 PM, Stellan Lagerström 
wrote:

> User toleary made some kind of upload last week of 200k+ nodes in
> Delaware, many of which seem to be duplicates of existing nodes:
> http://www.openstreetmap.org/user/toleary/history#map=9/39.0874/-75.5759
>
> Have sent message to user, but without answer.
>
> /Stellan
>
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Duplicate nodes in Delaware

[Paul Norman]

On 5/11/2015 3:47 PM, Bryce Nesbitt wrote:

Looks pretty bad.
Bare untagged nodes on top of existing way nodes.
A speedy revert is in order: are you comfortable doing that, or do you 
want help?
This is currently being reverted. It may take some time, given the size 
of the changesets.


___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Duplicate nodes in Delaware

[Bryce Nesbitt]

For those who don't know:

a speedy revert is better: if someone starts cleaning this import up
manually, it can create conflicts that
are much harder to resolve.

Thanks Paul!


On Mon, May 11, 2015 at 3:57 PM, Paul Norman  wrote:

> On 5/11/2015 3:47 PM, Bryce Nesbitt wrote:
>
>> Looks pretty bad.
>> Bare untagged nodes on top of existing way nodes.
>> A speedy revert is in order: are you comfortable doing that, or do you
>> want help?
>>
> This is currently being reverted. It may take some time, given the size of
> the changesets.
>
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] New York, Ellis Island Boundary

[stevea]

Jim McAndrew  writes:
This is correct, the island is part of NY, but the water is part of 
NJ, when they made the island bigger, the new land was in NJ.


Similar to how Delaware has the entire Delaware Bay in its 
boundaries. They added some fill to the NJ side, and now there's a 
piece of Delaware connected to New Jersey:

http://www.openstreetmap.org/#map=14/39.6142/-75.5635
http://en.wikipedia.org/wiki/Finns_Point


Might explain how OpenCycleMap is off in spots at z=9 with state 
names at Pennsylvania/Maryland/Delaware (Delaware is a long way from 
Cumberland).  I'm not aware of the history or explanation of that, so 
perhaps I could have better read up on bug reports, Andy (Allan).


There are little messes in our map.  Yet, we do achieve harmony and 
consensus.  Even with all the skipping of that middle there, I 
continue to find OSM a most amazing project.


(Like, the way Bryce just stood up to help out now:  nice).

Stiching up admin_level=4 boundaries, ah, yup.  Quite a map we're 
knitting here.


SteveA
California___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] New York, Ellis Island Boundary

[Kevin Kenny]

On 05/11/2015 08:37 PM, stevea wrote:

Re: [Talk-us] New York, Ellis Island Boundary
Jim McAndrew  writes:
This is correct, the island is part of NY, but the water is part of 
NJ, when they made the island bigger, the new land was in NJ.


Similar to how Delaware has the entire Delaware Bay in its 
boundaries. They added some fill to the NJ side, and now there's a 
piece of Delaware connected to New Jersey:

http://www.openstreetmap.org/#map=14/39.6142/-75.5635
http://en.wikipedia.org/wiki/Finns_Point


Might explain how OpenCycleMap is off in spots at z=9 with state names 
at Pennsylvania/Maryland/Delaware (Delaware is a long way from 
Cumberland).  I'm not aware of the history or explanation of that, so 
perhaps I could have better read up on bug reports, Andy (Allan).


I suspect that what's messing that up is the small spit of New Jersey 
north and west of Oldmans Creek.
http://kbk.is-a-geek.net/catskills/karl.html?la=39.7999&lo=-75.3995&z=13 
shows it in enough detail to show how lettering on the boundary would be 
tricky to render.


--
73 de ke9tv/2, Kevin


___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

[Talk-us] NYC High Line is Wonky on OSM

[Elliott Plack]

Friends,

I was attempting to do some pedestrian routing on the High Line (the
elevated park in NYC, see Wikipedia for background) and noticed some
oddities about how it was mapped on OSM. Quickly, this is a former elevated
train viaduct that has been converted into a popular park in Manhattan.
Since this is a popular area, I thought I'd ask the community first. Things
I've noticed:

1. There is a 'building=yes' way for the entire elevated portion, including
many of the supports that hold the platform up. This is pretty cool, and
probably looks neat in 3D. There are some building overlaps, where the line
goes through some buildings. https://www.openstreetmap.org/way/37054313
1a. The building also has the park tagging, which doesn't show up on the
map when tagged to the same way (apparently).
2. There are two parallel ways on the northern part of the park, one for
the former railway, another for the path. I believe that these should be
merged or at least share points. The former railway IS the pedestrian path,
so no need for parallel ways, right?
https://www.openstreetmap.org/way/46481094
2a. https://www.openstreetmap.org/way/305761607
3. Stairs like this should connect to the street.
https://www.openstreetmap.org/way/305761606
4. The 'highway=pedestrian' portion is not tagged as a bridge, which it is,
arguably. But then, if the viaduct is a 'building', is it actually a
bridge? I think it should be tagged as a bridge for cartography purposes.
4a. The 'highway=pedestrian' way does not have a name. The building does,
but that doesn't render well. Named ways should be named, right?
5. There are several 'highway=pedestrian' areas like this one. Is there a
better tag for open space like this?
https://www.openstreetmap.org/way/277945794
6. Things get really crazy with the building passages.
https://www.openstreetmap.org/way/277885773
7. There are a few oddities about the paths extending out from this node,
all these crossing ways are hard to comprehend.
https://www.openstreetmap.org/node/2823299563

Local mappers have clearly spent a lot of time on this, anyone have any
feedback about how this could be mapped better, if at all?


Best,

Elliott
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] NYC High Line is Wonky on OSM

[Brad Neuhauser]

Seems like it might be better to tag it as man_made=bridge rather than
building=*

http://wiki.openstreetmap.org/wiki/Tag:man_made%3Dbridge

On Mon, May 11, 2015 at 10:32 PM, Elliott Plack 
wrote:

> Friends,
>
> I was attempting to do some pedestrian routing on the High Line (the
> elevated park in NYC, see Wikipedia for background) and noticed some
> oddities about how it was mapped on OSM. Quickly, this is a former elevated
> train viaduct that has been converted into a popular park in Manhattan.
> Since this is a popular area, I thought I'd ask the community first. Things
> I've noticed:
>
> 1. There is a 'building=yes' way for the entire elevated portion,
> including many of the supports that hold the platform up. This is pretty
> cool, and probably looks neat in 3D. There are some building overlaps,
> where the line goes through some buildings.
> https://www.openstreetmap.org/way/37054313
> 1a. The building also has the park tagging, which doesn't show up on the
> map when tagged to the same way (apparently).
> 2. There are two parallel ways on the northern part of the park, one for
> the former railway, another for the path. I believe that these should be
> merged or at least share points. The former railway IS the pedestrian path,
> so no need for parallel ways, right?
> https://www.openstreetmap.org/way/46481094
> 2a. https://www.openstreetmap.org/way/305761607
> 3. Stairs like this should connect to the street.
> https://www.openstreetmap.org/way/305761606
> 4. The 'highway=pedestrian' portion is not tagged as a bridge, which it
> is, arguably. But then, if the viaduct is a 'building', is it actually a
> bridge? I think it should be tagged as a bridge for cartography purposes.
> 4a. The 'highway=pedestrian' way does not have a name. The building does,
> but that doesn't render well. Named ways should be named, right?
> 5. There are several 'highway=pedestrian' areas like this one. Is there a
> better tag for open space like this?
> https://www.openstreetmap.org/way/277945794
> 6. Things get really crazy with the building passages.
> https://www.openstreetmap.org/way/277885773
> 7. There are a few oddities about the paths extending out from this node,
> all these crossing ways are hard to comprehend.
> https://www.openstreetmap.org/node/2823299563
>
> Local mappers have clearly spent a lot of time on this, anyone have any
> feedback about how this could be mapped better, if at all?
>
>
> Best,
>
> Elliott
>
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us
>
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Duplicate nodes in Delaware

[Frederik Ramm]

Hi,

On 05/12/2015 12:57 AM, Paul Norman wrote:
> This is currently being reverted. It may take some time, given the size 
> of the changesets.

Reverted in changesets 31020612, 31021138, 31022607, 31022118, 31021597,
31022951.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frede...@remote.org  ##  N49°00'09" E008°23'33"

___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Duplicate nodes in Delaware

[Frederik Ramm]

Hi,

On 05/12/2015 01:13 AM, Bryce Nesbitt wrote:
> a speedy revert is better: if someone starts cleaning this import up
> manually, it can create conflicts that are much harder to resolve.

There were a couple "this node has already been deleted" situations
during the revert process but that's not a big problem. It is more
difficult if people start using badly-imported stuff as building blocks
for their own work because then it becomes hard to determine
automatically whether something is still revert-worthy or whether it has
been elevated to the rank of useful data by the mapper who touched it.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frede...@remote.org  ##  N49°00'09" E008°23'33"

___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us